JP5257551B1 - Exhaust gas purification device for internal combustion engine - Google Patents

Abstract

In the internal combustion engine, an upstream air-fuel ratio sensor (23), a hydrocarbon supply valve (15), an exhaust purification catalyst (13), and a downstream air-fuel ratio sensor (24) are arranged in order from the upstream in the engine exhaust passage. When hydrocarbons are supplied from the hydrocarbon supply valve (15), the air-fuel ratio detected by the downstream air-fuel ratio sensor (24) is detected when no hydrocarbon is supplied from the hydrocarbon supply valve (15). It changes to the rich side with respect to the reference air-fuel ratio. The amount of hydrocarbons supplied from the hydrocarbon supply valve (15) and passing through the exhaust purification catalyst (13) is detected by the air-fuel ratio detected by the upstream air-fuel ratio sensor (23) and the downstream air-fuel ratio sensor (24). It is detected from the air-fuel ratio difference from the reference air-fuel ratio.

Description

  The present invention relates to an exhaust emission control device for an internal combustion engine.

The engine exhaust passage, NO _X storage catalyst air-fuel ratio of the inflowing exhaust gas when the lean that releases NO _X air-fuel ratio of the exhaust gas which is occluded becomes rich for occluding NO _X contained in the exhaust gas inflow the place, the fuel addition valve disposed in the NO _X storage catalyst in the engine exhaust passage upstream of the air-fuel ratio sensor arranged in _the NO _X storage catalyst downstream of the engine exhaust passage, to release the NO _X from the NO _X storing catalyst it is known internal combustion engines, which the air-fuel ratio of the exhaust gas flowing into the NO _X storage catalyst by supplying fuel to the engine exhaust passage from the fuel supply valve to rich when it should (for example, see Patent Document 1).
By the way, in this internal combustion engine, when an excessive amount of excess hydrocarbons is supplied from the fuel addition valve, the amount of hydrocarbons that pass through the NO _X storage catalyst increases. In this case, if the amount of passing through of the hydrocarbon can be detected, the amount of the hydrocarbon consumed in vain can be reduced. However, the current situation is that such a slip-through amount of hydrocarbons cannot be detected by a simple method.

JP 2004-316458 A

  An object of the present invention is to provide an exhaust emission control device for an internal combustion engine that can detect the amount of passing through hydrocarbons by a simple method.

According to the present invention, a hydrocarbon supply valve for supplying hydrocarbons is disposed in the engine exhaust passage, and an upstream side empty space for detecting the air-fuel ratio of the exhaust gas in the engine exhaust passage upstream of the hydrocarbon supply valve. the ratio sensor disposed, an exhaust purification catalyst for reacting NO _X and reformed hydrocarbons contained in the hydrocarbon feed valve downstream of the engine exhaust passage in the exhaust gas is arranged, the exhaust gas purifying catalyst downstream A downstream air-fuel ratio sensor for detecting the air-fuel ratio of the exhaust gas is disposed in the engine exhaust passage, and the downstream air-fuel ratio sensor covers the solid electrolyte, the electrodes covering both sides of the solid electrolyte, and one of the electrodes. It has a diffusion resistance layer and a sensor of a type in which exhaust gas is guided onto the diffusion resistance layer. A noble metal catalyst is supported on the exhaust gas flow surface of the exhaust purification catalyst and a basic around the noble metal catalyst. Exhaust gas distribution A surface portion is formed, and when the exhaust purification catalyst vibrates the concentration of hydrocarbons flowing into the exhaust purification catalyst with an amplitude within a predetermined range and a period within a predetermined range, which has a property for reducing the NO _X contained, and storage amount of NO _X contained in the exhaust gas to be longer than the predetermined range vibration period of the hydrocarbon concentration has a property of increasing, carbonized When hydrocarbons are supplied from the hydrogen supply valve, the air-fuel ratio detected by the downstream air-fuel ratio sensor changes to the rich side with respect to the reference air-fuel ratio detected when no hydrocarbon is supplied from the hydrocarbon supply valve. The amount of hydrocarbons supplied from the hydrocarbon supply valve and passing through the exhaust purification catalyst is determined based on the air-fuel ratio detected by the upstream air-fuel ratio sensor and the reference air-fuel detected by the downstream air-fuel ratio sensor. Exhaust gas purification apparatus for an internal combustion engine so as to detect the air-fuel ratio difference is provided with.

  The amount of passing through hydrocarbons can be detected from the output signals of the upstream air-fuel ratio sensor and the downstream air-fuel ratio sensor.

FIG. 1 is an overall view of a compression ignition type internal combustion engine.
FIG. 2 is a view schematically showing the surface portion of the catalyst carrier.
FIG. 3 is a view for explaining an oxidation reaction in the exhaust purification catalyst.
FIG. 4 is a diagram showing changes in the air-fuel ratio of the exhaust gas flowing into the exhaust purification catalyst.
Figure 5 is a diagram illustrating a NO _X purification rate.
6A and 6B are diagrams for explaining the oxidation-reduction reaction in the exhaust purification catalyst.
7A and 7B are diagrams for explaining the oxidation-reduction reaction in the exhaust purification catalyst.
FIG. 8 is a diagram showing a change in the air-fuel ratio of the exhaust gas flowing into the exhaust purification catalyst.
Figure 9 is a diagram illustrating a NO _X purification rate.
FIG. 10 is a time chart showing changes in the air-fuel ratio of the exhaust gas flowing into the exhaust purification catalyst.
FIG. 11 is a time chart showing changes in the air-fuel ratio of the exhaust gas flowing into the exhaust purification catalyst.
FIG. 12 is a diagram showing the relationship between the oxidizing power of the exhaust purification catalyst and the required minimum air-fuel ratio X.
FIG. 13 is a graph showing the relationship between the oxygen concentration in the exhaust gas and the amplitude ΔH of the hydrocarbon concentration, with which the same NO _x purification rate can be obtained.
Figure 14 is a diagram showing the relationship between the amplitude ΔH and NO _X purification rate of hydrocarbon concentration.
Figure 15 is a diagram showing the relationship between the vibration period ΔT and NO _X purification rate of hydrocarbon concentration.
FIG. 16 is a diagram showing changes in the air-fuel ratio of exhaust gas flowing into the exhaust purification catalyst.
Figure 17 is a diagram illustrating a map of exhaust amount of NO _X NOXA.
FIG. 18 is a diagram showing the fuel injection timing.
FIG. 19 is a diagram showing a map of the additional fuel amount WR.
20A and 20B are diagrams showing the target base air-fuel ratio.
21A, 21B, and 21C are diagrams showing a hydrocarbon injection period and the like.
22A and 22B are time charts showing changes in the air-fuel ratio of the exhaust gas flowing into the exhaust purification catalyst.
FIG. 23 is a diagram showing a map of the target peak air-fuel ratio.
24A and 24B are diagrams showing the structure of the air-fuel ratio sensor shown schematically.
25A and 25B are diagrams showing changes in the air-fuel ratio detected by the downstream air-fuel ratio sensor.
26A and 26B are diagrams showing changes in the air-fuel ratio detected by the downstream air-fuel ratio sensor and the like.
FIG. 27 is a flowchart for performing operation control.
28 and 29 are flowcharts showing an embodiment of the operation control I.
30A and 30B are views showing changes in the air-fuel ratio detected by the downstream air-fuel ratio sensor.
31 and 32 are flowcharts showing another embodiment of the operation control I.
FIG. 33 is a flowchart for executing flag control.

  FIG. 1 shows an overall view of a compression ignition type internal combustion engine.
  Referring to FIG. 1, 1 is an engine body, 2 is a combustion chamber of each cylinder, 3 is an electronically controlled fuel injection valve for injecting fuel into each combustion chamber 2, 4 is an intake manifold, and 5 is an exhaust manifold. Respectively. The intake manifold 4 is connected to the outlet of the compressor 7 a of the exhaust turbocharger 7 via the intake duct 6, and the inlet of the compressor 7 a is connected to the air cleaner 9 via the intake air amount detector 8. A throttle valve 10 driven by a step motor is disposed in the intake duct 6, and a cooling device 11 for cooling intake air flowing through the intake duct 6 is disposed around the intake duct 6. In the embodiment shown in FIG. 1, the engine cooling water is guided into the cooling device 11, and the intake air is cooled by the engine cooling water.
  On the other hand, the exhaust manifold 5 is connected to the inlet of the exhaust turbine 7 b of the exhaust turbocharger 7. The outlet of the exhaust turbine 7b is connected to the inlet of the exhaust purification catalyst 13 via the exhaust pipe 12a, and the outlet of the exhaust purification catalyst 13 is used to collect particulates contained in the exhaust gas via the exhaust pipe 12b. It is connected to the particulate filter 14.
  In the exhaust pipe 12a upstream of the exhaust purification catalyst 13, a hydrocarbon supply valve 15 for supplying hydrocarbons made of light oil or other fuel used as fuel for the compression ignition internal combustion engine is arranged. In the embodiment shown in FIG. 1, light oil is used as the hydrocarbon supplied from the hydrocarbon supply valve 15. The present invention can also be applied to a spark ignition type internal combustion engine in which combustion is performed under a lean air-fuel ratio. In this case, the hydrocarbon supply valve 15 supplies hydrocarbons made of gasoline or other fuel used as fuel for the spark ignition internal combustion engine.
  On the other hand, the exhaust manifold 5 and the intake manifold 4 are connected to each other via an exhaust gas recirculation (hereinafter referred to as EGR) passage 16, and an electronically controlled EGR control valve 17 is disposed in the EGR passage 16. A cooling device 18 for cooling the EGR gas flowing in the EGR passage 16 is disposed around the EGR passage 16. In the embodiment shown in FIG. 1, the engine cooling water is guided into the cooling device 18, and the EGR gas is cooled by the engine cooling water. On the other hand, each fuel injection valve 3 is connected to a common rail 20 via a fuel supply pipe 19, and this common rail 20 is connected to a fuel tank 22 via an electronically controlled fuel pump 21 having a variable discharge amount. The fuel stored in the fuel tank 22 is supplied into the common rail 20 by the fuel pump 21, and the fuel supplied into the common rail 20 is supplied to the fuel injection valve 3 through each fuel supply pipe 19.
  The electronic control unit 30 is composed of a digital computer, and is connected to each other by a bidirectional bus 31. A ROM (read only memory) 32, a RAM (random access memory) 33, a CPU (microprocessor) 34, an input port 35 and an output port 36. It comprises. An upstream air-fuel ratio sensor 23 for detecting the air-fuel ratio of the exhaust gas discharged from the engine is disposed in the exhaust pipe 12 a upstream of the hydrocarbon supply valve 15, and the exhaust pipe 12 b downstream of the exhaust purification catalyst 13. A downstream air-fuel ratio sensor 24 for detecting the air-fuel ratio of the exhaust gas flowing out from the exhaust purification catalyst 13 is disposed inside. Further, a temperature sensor 25 for detecting the temperature of the exhaust purification catalyst 13 is disposed downstream of the exhaust purification catalyst 13, and the particulate filter 14 is for detecting a differential pressure before and after the particulate filter 14. A differential pressure sensor 26 is attached. Output signals of the upstream air-fuel ratio sensor 23, downstream air-fuel ratio sensor 24, temperature sensor 25, differential pressure sensor 26, and intake air amount detector 8 are input to the input port 35 via corresponding AD converters 37, respectively. The
  A load sensor 41 that generates an output voltage proportional to the depression amount L of the accelerator pedal 40 is connected to the accelerator pedal 40, and the output voltage of the load sensor 41 is input to the input port 35 via the corresponding AD converter 37. Is done. Further, the input port 35 is connected to a crank angle sensor 42 that generates an output pulse every time the crankshaft rotates, for example, 15 °. On the other hand, the output port 36 is connected to the fuel injection valve 3, the step motor for driving the throttle valve 10, the hydrocarbon supply valve 15, the EGR control valve 17, and the fuel pump 21 through corresponding drive circuits 38.
  FIG. 2 schematically shows the surface portion of the catalyst carrier carried on the substrate of the exhaust purification catalyst 13. In this exhaust purification catalyst 13, as shown in FIG. 2, noble metal catalysts 51 and 52 are supported on a catalyst support 50 made of alumina, for example, and further on this catalyst support 50 potassium K, sodium Na, cesium Cs. Alkaline metals such as barium Ba, alkaline earth metals such as calcium Ca, rare earths such as lanthanoids and silver Ag, copper Cu, iron Fe, NO such as iridium Ir_XA basic layer 53 containing at least one selected from metals capable of donating electrons is formed. Since the exhaust gas flows along the catalyst carrier 50, it can be said that the noble metal catalysts 51 and 52 are supported on the exhaust gas flow surface of the exhaust purification catalyst 13. Further, since the surface of the basic layer 53 is basic, the surface of the basic layer 53 is referred to as a basic exhaust gas flow surface portion 54.
  On the other hand, in FIG. 2, the noble metal catalyst 51 is made of platinum Pt, and the noble metal catalyst 52 is made of rhodium Rh. That is, the noble metal catalysts 51 and 52 carried on the catalyst carrier 50 are composed of platinum Pt and rhodium Rh. In addition to platinum Pt and rhodium Rh, palladium Pd can be further supported on the catalyst carrier 50 of the exhaust purification catalyst 13, or palladium Pd can be supported instead of rhodium Rh. That is, the noble metal catalysts 51 and 52 supported on the catalyst carrier 50 are composed of platinum Pt and at least one of rhodium Rh and palladium Pd.
  When hydrocarbons are injected into the exhaust gas from the hydrocarbon supply valve 15, the hydrocarbons are reformed in the exhaust purification catalyst 13. In the present invention, NO is used in the exhaust purification catalyst 13 by using the reformed hydrocarbon at this time._XTo purify. FIG. 3 schematically shows the reforming action performed in the exhaust purification catalyst 13 at this time. As shown in FIG. 3, the hydrocarbon HC injected from the hydrocarbon feed valve 15 is converted into a radical hydrocarbon HC having a small number of carbons by the catalyst 51.
  FIG. 4 shows the supply timing of hydrocarbons from the hydrocarbon supply valve 15 and changes in the air-fuel ratio (A / F) in of the exhaust gas flowing into the exhaust purification catalyst 13. Since the change in the air-fuel ratio (A / F) in depends on the change in the concentration of hydrocarbons in the exhaust gas flowing into the exhaust purification catalyst 13, the air-fuel ratio (A / F) in shown in FIG. It can be said that the change represents a change in hydrocarbon concentration. However, since the air-fuel ratio (A / F) in decreases as the hydrocarbon concentration increases, the hydrocarbon concentration increases as the air-fuel ratio (A / F) in becomes richer in FIG.
  FIG. 5 shows a change in the air-fuel ratio (A / F) in of the exhaust gas flowing into the exhaust purification catalyst 13 as shown in FIG. 4 by periodically changing the concentration of hydrocarbons flowing into the exhaust purification catalyst 13. NO by the exhaust purification catalyst 13 when_XThe purification rate is shown for each catalyst temperature TC of the exhaust purification catalyst 13. The inventor has NO over a long period of time._XIn the research course, when the concentration of hydrocarbons flowing into the exhaust purification catalyst 13 is vibrated with an amplitude within a predetermined range and a period within a predetermined range, FIG. As shown in Fig. 4, extremely high NO even in a high temperature region of 400 ° C or higher._XIt has been found that a purification rate can be obtained.
  Further, at this time, a large amount of the reducing intermediate containing nitrogen and hydrocarbon continues to be held or adsorbed on the surface of the basic layer 53, that is, on the basic exhaust gas flow surface portion 54 of the exhaust purification catalyst 13. Reducing intermediate is high NO_XIt turns out that it plays a central role in obtaining the purification rate. Next, this will be described with reference to FIGS. 6A and 6B. 6A and 6B schematically show the surface portion of the catalyst carrier 50 of the exhaust purification catalyst 13, and in these FIGS. 6A and 6B, the concentration of hydrocarbons flowing into the exhaust purification catalyst 13 is predetermined. The reaction is shown to be presumed to occur when oscillated with an amplitude within a range and a period within a predetermined range.
  FIG. 6A shows a case where the concentration of hydrocarbons flowing into the exhaust purification catalyst 13 is low, and FIG. 6B shows that the concentration of hydrocarbons flowing into the exhaust purification catalyst 13 when hydrocarbons are supplied from the hydrocarbon supply valve 15 is high. It shows when
  As can be seen from FIG. 4, since the air-fuel ratio of the exhaust gas flowing into the exhaust purification catalyst 13 is maintained lean except for a moment, the exhaust gas flowing into the exhaust purification catalyst 13 is usually in an oxygen excess state. Therefore, NO contained in the exhaust gas is oxidized on the platinum 51 as shown in FIG.₂And then this NO₂Is further oxidized to NO₃It becomes. NO₂Part of is NO₂ ⁻It becomes. In this case, NO₃The amount of production is NO₂ ⁻Much more than the amount of product. Therefore, a large amount of NO is present on platinum Pt51.₃And a small amount of NO₂ ⁻Will be generated. These NO₃And NO₂ ⁻Is highly active, and these NO₃And NO₂ ⁻Active NO_X ^*Called.
  On the other hand, when hydrocarbons are supplied from the hydrocarbon supply valve 15, as shown in FIG. 3, the hydrocarbons are reformed in the exhaust purification catalyst 13 and become radicals. As a result, as shown in FIG._X ^*The surrounding hydrocarbon concentration increases. By the way, active NO_X ^*After NO is generated, active NO_X ^*If the surrounding oxygen concentration is high for more than a certain period of time, active NO_X ^*Is oxidized and nitrate ion NO₃ ⁻In the basic layer 53. However, the active NO_X ^*When the surrounding hydrocarbon concentration is increased, as shown in FIG._X ^*Reacts with radical hydrocarbons HC on platinum 51, thereby producing a reducing intermediate. This reducing intermediate is attached or adsorbed on the surface of the basic layer 53.
  Note that the first reducing intermediate produced at this time is the nitro compound R-NO.₂It is thought that. This nitro compound R-NO₂Is produced, it becomes a nitrile compound R-CN, but this nitrile compound R-CN can only survive for a moment in that state, so it immediately becomes an isocyanate compound R-NCO. When this isocyanate compound R-NCO is hydrolyzed, the amine compound R-NH₂It becomes. However, in this case, it is considered that a part of the isocyanate compound R-NCO is hydrolyzed. Therefore, as shown in FIG. 6B, most of the reducing intermediates retained or adsorbed on the surface of the basic layer 53 are the isocyanate compound R-NCO and the amine compound R-NH.₂It is thought that.
  On the other hand, as shown in FIG. 6B, when hydrocarbon HC surrounds the generated reducing intermediate, the reducing intermediate is blocked by hydrocarbon HC and the reaction does not proceed further. In this case, the concentration of hydrocarbons flowing into the exhaust purification catalyst 13 is lowered, and as a result, when the oxygen concentration is increased, the hydrocarbons around the reducing intermediate are oxidized. As a result, as shown in FIG. 6A, the reducing intermediate and the active NO_X ^*Will react. Active NO at this time_X ^*Is a reducing intermediate R-NCO or R-NH₂Reacts with N₂, CO₂, H₂O, so NO_XWill be purified.
  In this way, in the exhaust purification catalyst 13, a reducing intermediate is generated by increasing the concentration of hydrocarbons flowing into the exhaust purification catalyst 13, and the concentration of hydrocarbons flowing into the exhaust purification catalyst 13 is decreased to reduce the oxygen concentration. By increasing the active NO_X ^*Reacts with reducing intermediates and NO_XIs purified. That is, the exhaust purification catalyst 13 makes NO._XIn order to purify, it is necessary to periodically change the concentration of hydrocarbons flowing into the exhaust purification catalyst 13.
  Of course, in this case, it is necessary to increase the concentration of the hydrocarbon to a concentration sufficiently high to produce a reducing intermediate,_X ^*It is necessary to reduce the hydrocarbon concentration to a concentration low enough to react with. That is, it is necessary to vibrate the concentration of hydrocarbons flowing into the exhaust purification catalyst 13 with an amplitude within a predetermined range. In this case, the generated reducing intermediate is active NO._X ^*Sufficient amounts of reducing intermediates R-NCO and R-NH until₂Must be retained on the basic layer 53, that is, on the basic exhaust gas flow surface portion 54. For this purpose, a basic exhaust gas flow surface portion 54 is provided.
  On the other hand, if the supply cycle of the hydrocarbon is lengthened, the period during which the oxygen concentration becomes high after the hydrocarbon is supplied and before the next hydrocarbon is supplied becomes longer, so that the active NO._X ^*Is absorbed in the basic layer 53 in the form of nitrate without producing a reducing intermediate. In order to avoid this, it is necessary to oscillate the concentration of hydrocarbons flowing into the exhaust purification catalyst 13 with a period within a predetermined range.
  Therefore, in the embodiment according to the present invention, NO contained in the exhaust gas._XReductive intermediates R-NCO and R-NH containing nitrogen and hydrocarbons by reacting with modified hydrocarbons₂In order to generate NO, noble metal catalysts 51 and 52 are supported on the exhaust gas flow surface of the exhaust purification catalyst 13, and the generated reducing intermediates R-NCO and R-NH₂Is maintained around the noble metal catalyst 51, 52, a basic exhaust gas flow surface portion 54 is formed around the noble metal catalyst 51, 52, and is held on the basic exhaust gas flow surface portion 54. Reducing intermediates R-NCO and R-NH₂NO_XIs reduced, and the vibration period of the hydrocarbon concentration is reduced by reducing intermediates R-NCO and R-NH.₂Is the oscillation period necessary to continue to generate Incidentally, in the example shown in FIG. 4, the injection interval is 3 seconds.
  When the oscillation period of the hydrocarbon concentration, that is, the supply period of the hydrocarbon HC is longer than the period within the above-mentioned predetermined range, the reducing intermediates R-NCO and R-NH are formed from the surface of the basic layer 53.₂Disappears, and at this time, the active NO produced on platinum Pt53_X ^*Is nitrate ion NO as shown in FIG. 7A.₃ ⁻In the form of nitrate in the form of nitrate. That is, at this time, NO in the exhaust gas_XWill be absorbed in the basic layer 53 in the form of nitrate.
  On the other hand, FIG._XThis shows a case where the air-fuel ratio of the exhaust gas flowing into the exhaust purification catalyst 13 is made the stoichiometric air-fuel ratio or rich when NO is absorbed in the basic layer 53 in the form of nitrate. In this case, the reaction proceeds in the reverse direction (NO₃ ⁻→ NO₂), And thus the nitrates absorbed in the basic layer 53 are successively converted to nitrate ions NO.₃ ⁻And NO as shown in FIG. 7B₂From the basic layer 53. Then released NO₂Is reduced by hydrocarbons HC and CO contained in the exhaust gas.
  FIG. 8 shows the NO of the basic layer 53_XThis shows a case where the air-fuel ratio (A / F) in of the exhaust gas flowing into the exhaust purification catalyst 13 is temporarily made rich slightly before the absorption capacity is saturated. In the example shown in FIG. 8, the time interval of this rich control is 1 minute or more. In this case, NO absorbed in the basic layer 53 when the air-fuel ratio (A / F) in of the exhaust gas is lean._XIs released from the basic layer 53 and reduced when the air-fuel ratio (A / F) in of the exhaust gas is temporarily made rich. Therefore, in this case, the basic layer 53 is NO._XIt plays the role of an absorbent for temporarily absorbing.
  At this time, the basic layer 53 is NO._XTherefore, if the term occlusion is used as a term including both absorption and adsorption, the basic layer 53 is not NO at this time._XNO for temporary storage_XIt plays the role of a storage agent. That is, in this case, if the ratio of air and fuel (hydrocarbon) supplied into the engine intake passage, the combustion chamber 2 and the exhaust passage upstream of the exhaust purification catalyst 13 is referred to as the air-fuel ratio of the exhaust gas, the exhaust purification catalyst. 13 is NO when the air-fuel ratio of the exhaust gas is lean_XNO is stored when the oxygen concentration in the exhaust gas decreases._XNO release_XIt functions as a storage catalyst.
  FIG. 9 shows that the exhaust purification catalyst 13 is thus NO._XNO when functioning as a storage catalyst_XThe purification rate is shown. The horizontal axis in FIG. 9 indicates the catalyst temperature TC of the exhaust purification catalyst 13. Set the exhaust purification catalyst 13 to NO_XWhen functioning as an occlusion catalyst, as shown in FIG. 9, when the catalyst temperature TC is 300 ° C. to 400 ° C., extremely high NO_XA purification rate can be obtained, but NO when the catalyst temperature TC reaches 400 ° C or higher._XThe purification rate decreases.
  Thus, when the catalyst temperature TC reaches 400 ° C. or higher, NO_XThe purification rate decreases because when the catalyst temperature TC reaches 400 ° C. or higher, the nitrate is thermally decomposed and NO.₂This is because it is discharged from the exhaust purification catalyst 13 in the form of. That is, NO_XAs long as the catalyst temperature TC is high._XIt is difficult to obtain a purification rate. However, the new NO shown in FIGS. 4 to 6A and 6B_XIn the purification method, as can be seen from FIGS. 6A and 6B, nitrate is not generated or is very small even if it is generated. Therefore, even when the catalyst temperature TC is high as shown in FIG._XA purification rate will be obtained.
  Therefore, in the present invention, the hydrocarbon supply valve 15 for supplying hydrocarbons is arranged in the engine exhaust passage, and NO contained in the exhaust gas in the engine exhaust passage downstream of the hydrocarbon supply valve 15._XAn exhaust purification catalyst 13 for reacting the catalyst with the reformed hydrocarbon is disposed. Noble metal catalysts 51 and 52 are supported on the exhaust gas flow surface of the exhaust purification catalyst 13 and around the noble metal catalysts 51 and 52. Is formed with a basic exhaust gas flow surface portion 54, and the exhaust purification catalyst 13 sets the concentration of hydrocarbons flowing into the exhaust purification catalyst 13 within a predetermined range and an amplitude within a predetermined range. When it is vibrated with a period of_XNO and contained in the exhaust gas when the oscillation period of the hydrocarbon concentration is longer than this predetermined range._XAnd the hydrocarbon concentration flowing into the exhaust purification catalyst 13 during engine operation is vibrated with an amplitude within a predetermined range and a period within a predetermined range, As a result, NO contained in the exhaust gas_XIs reduced in the exhaust purification catalyst 13.
  That is, the NO shown in FIGS. 4 to 6A and 6B._XThe purification method carries a noble metal catalyst and NO._XIn the case of using an exhaust purification catalyst having a basic layer capable of absorbing NO, NO hardly forms nitrates._XNew NO to purify_XIt can be said that it is a purification method. In fact, this new NO_XWhen the purification method is used, the exhaust purification catalyst 13 is set to NO._XCompared with the case where it functions as an occlusion catalyst, the amount of nitrate detected from the basic layer 53 is extremely small. This new NO_XThe purification method is hereinafter referred to as the first NO._XThis is called a purification method.
  Next, referring to FIGS. 10 to 15, the first NO._XThe purification method will be described in a little more detail.
  FIG. 10 shows an enlarged view of the change in the air-fuel ratio (A / F) in shown in FIG. As described above, the change in the air-fuel ratio (A / F) in of the exhaust gas flowing into the exhaust purification catalyst 13 indicates the change in the concentration of hydrocarbons flowing into the exhaust purification catalyst 13 at the same time. In FIG. 10, ΔH indicates the amplitude of the change in the concentration of hydrocarbon HC flowing into the exhaust purification catalyst 13, and ΔT indicates the oscillation period of the concentration of hydrocarbon flowing into the exhaust purification catalyst 13.
  Further, in FIG. 10, (A / F) b represents the base air-fuel ratio indicating the air-fuel ratio of the combustion gas for generating the engine output. In other words, the base air-fuel ratio (A / F) b represents the air-fuel ratio of the exhaust gas flowing into the exhaust purification catalyst 13 when the supply of hydrocarbons is stopped. On the other hand, in FIG. 10, X represents the generated active NO._X ^*Represents the upper limit of the air-fuel ratio (A / F) in used for the production of the reducing intermediate without being occluded in the basic layer 53 in the form of nitrate,_X ^*It is necessary to make the air-fuel ratio (A / F) in lower than the upper limit X of this air-fuel ratio in order to cause the reduced hydrocarbon to react with the reformed hydrocarbon.
  In other words, X in FIG._X ^*Represents the lower limit of the hydrocarbon concentration required to produce a reducing intermediate by reacting the modified hydrocarbon with the modified hydrocarbon. It is necessary to make it higher than the lower limit X. In this case, whether or not a reducing intermediate is generated depends on the active NO._X ^*The ratio between the surrounding oxygen concentration and the hydrocarbon concentration, that is, the air-fuel ratio (A / F) in, is determined by the above-mentioned upper limit X of the air-fuel ratio necessary for generating the reducing intermediate, Called.
  In the example shown in FIG. 10, the required minimum air-fuel ratio X is rich. Therefore, in this case, the air-fuel ratio (A / F) in is instantaneously required to generate the reducing intermediate. The following is made rich: On the other hand, in the example shown in FIG. 11, the required minimum air-fuel ratio X is lean. In this case, the reducing intermediate is generated by periodically reducing the air-fuel ratio (A / F) in while maintaining the air-fuel ratio (A / F) in lean.
  In this case, whether the required minimum air-fuel ratio X becomes rich or lean depends on the oxidizing power of the exhaust purification catalyst 13. In this case, for example, if the amount of the precious metal 51 supported is increased, the exhaust purification catalyst 13 becomes stronger in oxidizing power, and if it becomes more acidic, the oxidizing power becomes stronger. Accordingly, the oxidizing power of the exhaust purification catalyst 13 varies depending on the amount of the precious metal 51 supported and the acidity.
  When the exhaust purification catalyst 13 having a strong oxidizing power is used, the air-fuel ratio (A / F) in is periodically decreased while maintaining the air-fuel ratio (A / F) in lean as shown in FIG. When the air-fuel ratio (A / F) in is lowered, the hydrocarbon is completely oxidized, and as a result, a reducing intermediate cannot be generated. On the other hand, when the exhaust purification catalyst 13 having a strong oxidizing power is used, if the air-fuel ratio (A / F) in is periodically made rich as shown in FIG. 10, the air-fuel ratio (A / F) in is rich. When selected, some hydrocarbons are partially oxidized without being completely oxidized, i.e., the hydrocarbon is reformed, thus producing a reducing intermediate. Therefore, when the exhaust purification catalyst 13 having a strong oxidizing power is used, the required minimum air-fuel ratio X needs to be made rich.
  On the other hand, when the exhaust purification catalyst 13 having a weak oxidizing power is used, the air-fuel ratio (A / F) in is periodically decreased while maintaining the air-fuel ratio (A / F) in lean as shown in FIG. As a result, some of the hydrocarbons are not completely oxidized but are partially oxidized, that is, the hydrocarbons are reformed, thus producing a reducing intermediate. On the other hand, when the exhaust purification catalyst 13 having a weak oxidizing power is used, if the air-fuel ratio (A / F) in is periodically made rich as shown in FIG. 10, a large amount of hydrocarbons are not oxidized. The exhaust gas is simply exhausted from the exhaust purification catalyst 13, and the amount of hydrocarbons that are wasted is increased. Therefore, when the exhaust purification catalyst 13 having a weak oxidizing power is used, the required minimum air-fuel ratio X needs to be made lean.
  That is, it can be seen that the required minimum air-fuel ratio X needs to be lowered as the oxidizing power of the exhaust purification catalyst 13 becomes stronger, as shown in FIG. As described above, the required minimum air-fuel ratio X becomes lean or rich due to the oxidizing power of the exhaust purification catalyst 13, but the case where the required minimum air-fuel ratio X is rich will be described as an example. The amplitude of the change in the concentration of the inflowing hydrocarbon and the oscillation period of the concentration of the hydrocarbon flowing into the exhaust purification catalyst 13 will be described.
  When the base air-fuel ratio (A / F) b increases, that is, when the oxygen concentration in the exhaust gas before the hydrocarbons are supplied increases, the air-fuel ratio (A / F) in is made equal to or less than the required minimum air-fuel ratio X. As a result, the amount of hydrocarbons necessary for the increase increases, and the amount of excess hydrocarbons that did not contribute to the production of the reducing intermediate also increases. In this case, NO_XAs described above, it is necessary to oxidize the surplus hydrocarbons in order to purify the water well._XIn order to purify the water well, a larger amount of excess hydrocarbon requires more oxygen.
  In this case, the amount of oxygen can be increased by increasing the oxygen concentration in the exhaust gas. Therefore NO_XIn order to purify the gas well, it is necessary to increase the oxygen concentration in the exhaust gas after the hydrocarbon is supplied when the oxygen concentration in the exhaust gas before the hydrocarbon is supplied is high. That is, it is necessary to increase the amplitude of the hydrocarbon concentration as the oxygen concentration in the exhaust gas before the hydrocarbon is supplied is higher.
  Figure 13 shows the same NO_XIt shows the relationship between the oxygen concentration in the exhaust gas before the hydrocarbon is supplied and the amplitude ΔH of the hydrocarbon concentration when the purification rate is obtained. The same NO from FIG._XIt can be seen that in order to obtain the purification rate, it is necessary to increase the amplitude ΔH of the hydrocarbon concentration as the oxygen concentration in the exhaust gas before the hydrocarbon is supplied is higher. That is, the same NO_XIn order to obtain the purification rate, it is necessary to increase the amplitude ΔT of the hydrocarbon concentration as the base air-fuel ratio (A / F) b increases. In other words, NO_XIn order to purify the gas well, the amplitude ΔT of the hydrocarbon concentration can be reduced as the base air-fuel ratio (A / F) b becomes lower.
  By the way, the base air-fuel ratio (A / F) b becomes the lowest during the acceleration operation. At this time, if the amplitude ΔH of the hydrocarbon concentration is about 200 ppm, NO_XCan be purified well. The base air-fuel ratio (A / F) b is usually larger than that during acceleration operation. Therefore, as shown in FIG. 14, when the hydrocarbon concentration amplitude ΔH is 200 ppm or more, good NO is obtained._XA purification rate can be obtained.
  On the other hand, when the base air-fuel ratio (A / F) b is the highest, if the amplitude ΔH of the hydrocarbon concentration is about 10000 ppm, good NO_XIt is known that a purification rate can be obtained. Therefore, in the present invention, the predetermined range of the amplitude of the hydrocarbon concentration is set to 200 ppm to 10,000 ppm.
  In addition, when the vibration period ΔT of the hydrocarbon concentration becomes longer, after the hydrocarbon is supplied, the active NO is maintained while the hydrocarbon is supplied next._X ^*The surrounding oxygen concentration becomes high. In this case, when the vibration period ΔT of the hydrocarbon concentration is longer than about 5 seconds, the active NO_X ^*Will begin to be absorbed in the basic layer 53 in the form of nitrate, and therefore, as shown in FIG. 15, when the vibration period ΔT of the hydrocarbon concentration becomes longer than about 5 seconds, NO_XThe purification rate will decrease. Therefore, the vibration period ΔT of the hydrocarbon concentration needs to be 5 seconds or less.
  On the other hand, when the vibration period ΔT of the hydrocarbon concentration becomes approximately 0.3 seconds or less, the supplied hydrocarbon starts to accumulate on the exhaust gas flow surface of the exhaust purification catalyst 13, and therefore, as shown in FIG. When the vibration period ΔT of the motor becomes approximately 0.3 seconds or less, NO_XThe purification rate decreases. Therefore, in the present invention, the vibration period of the hydrocarbon concentration is set to be between 0.3 seconds and 5 seconds.
  Next, referring to FIG. 16 to FIG._XNO when functioning as a storage catalyst_XThe purification method will be specifically described. In this way, the exhaust purification catalyst 13 is changed to NO._XNO when functioning as a storage catalyst_XThe purification method is hereinafter referred to as the second NO._XThis is called a purification method.
  This second NO_XIn the purification method, the occluded NO occluded in the basic layer 53 as shown in FIG._XWhen the amount ΣNOX exceeds a predetermined allowable amount MAX, the air-fuel ratio (A / F) in of the exhaust gas flowing into the exhaust purification catalyst 13 is temporarily made rich. When the air-fuel ratio (A / F) in of the exhaust gas is made rich, the NO stored in the basic layer 53 when the air-fuel ratio (A / F) in of the exhaust gas is lean_XAre released from the basic layer 53 at once and reduced. NO_XIs purified.
  Occlusion NO_XThe amount ΣNOX is, for example, NO discharged from the engine_XCalculated from the quantity. In the embodiment according to the present invention, the emission NO discharged from the engine per unit time_XThe amount NOXA is stored in advance in the ROM 32 as a function of the engine output torque Te and the engine speed N in the form of a map as shown in FIG._XFrom NOXA to NO_XAn amount ΣNOX is calculated. In this case, as described above, the period during which the air-fuel ratio (A / F) in of the exhaust gas is made rich is usually 1 minute or more.
  This second NO_XIn the purification method, as shown in FIG. 18, the air-fuel ratio of the exhaust gas flowing into the exhaust purification catalyst 13 by injecting the additional fuel WR in addition to the output generating fuel Q from the fuel injection valve 3 into the combustion chamber 2 ( A / F) in is made rich. The horizontal axis in FIG. 18 indicates the crank angle. This additional fuel WR is injected when it burns but does not appear as engine output, that is, slightly before ATDC 90 ° after compression top dead center. This fuel amount WR is stored in advance in the ROM 32 in the form of a map as shown in FIG. 19 as a function of the engine output torque Te and the engine speed N. Of course, the air-fuel ratio (A / F) in of the exhaust gas can be made rich by increasing the amount of hydrocarbons supplied from the hydrocarbon supply valve 15 in this case.
  Now again, the first NO_XReturning to the explanation of the purification method, the first NO_XNO using the purification method_XAs described above, it is necessary to appropriately control the amplitude ΔH and the vibration period ΔT of the hydrocarbon concentration. That is, the first NO_XNO using the purification method_XIn order to purify the gas well, the amplitude ΔH of the hydrocarbon concentration is controlled so that the air-fuel ratio (A / F) in of the exhaust gas flowing into the exhaust purification catalyst 13 is equal to or less than the required minimum air-fuel ratio X, It is necessary to control the oscillation period ΔT of the concentration between 0.3 seconds and 5 seconds.
  In this case, in the present invention, the amplitude ΔH of the hydrocarbon concentration is controlled to control the injection amount of hydrocarbons from the hydrocarbon feed valve 15, and the vibration period ΔT of the hydrocarbon concentration is controlled by the hydrocarbons from the hydrocarbon feed valve 15. It is controlled by controlling the injection cycle. In this case, the injection amount of hydrocarbons from the hydrocarbon supply valve 15 can be controlled by controlling at least one of the injection time or injection pressure of hydrocarbons from the hydrocarbon supply valve 15. However, hereinafter, the present invention will be described by taking as an example a case where the injection amount is controlled by controlling the injection time while keeping the injection pressure constant.
  In the embodiment according to the present invention, the optimum opening degree of the throttle valve 10 and the optimum opening degree of the EGR control valve 17 corresponding to the operating state of the engine are obtained in advance by experiments, and the throttle valve 10 and the EGR control are further determined. The optimum base air-fuel ratio (A / F) b obtained when the valve 17 is set to the optimum opening degree is also obtained in advance by experiments. FIG. 20A shows the optimum base air-fuel ratio (A / F) b obtained by this experiment as a function of the engine speed N and the engine output torque Te. Note that each solid line in FIG. 20A represents an air-fuel ratio equal air-fuel ratio line indicated by a numerical value.
  During engine operation, the fuel injection amount from the fuel injection valve 3 is controlled so that the air-fuel ratio of the exhaust gas discharged from the engine becomes the optimum base air-fuel ratio (A / F) b shown in FIG. 20A. The optimal base air-fuel ratio (A / F) b shown in FIG. 20A is stored in advance in the ROM 32 in the form of a map as shown in FIG. 20B as a function of the engine speed N and the engine output torque Te. Yes.
  Also, the first NO_XNO by purification method_XThe highest NO when the purification action is performed_XAn optimum hydrocarbon injection period ΔT and an optimum hydrocarbon injection time WT that can obtain a purification rate are obtained in advance by experiments. FIG. 21A shows the optimum hydrocarbon injection period ΔT obtained by this experiment as a function of the engine speed N and the engine output torque Te. In addition, each solid line in FIG. 21A indicates an equal hydrocarbon injection cycle. As can be seen from FIG. 21A, the optimum hydrocarbon injection period ΔT becomes shorter as the engine speed N becomes lower, and becomes shorter as the engine output torque Te becomes higher.
  The optimum hydrocarbon injection period ΔT shown in FIG. 21A is stored in advance in the ROM 32 in the form of a map as shown in FIG. 21B as a function of the engine speed N and the engine output torque Te. Further, the optimum hydrocarbon injection time WT obtained by experiment is also stored in advance in the ROM 32 in the form of a map as shown in FIG. 20C as a function of the engine speed N and the engine output torque Te.
  FIG. 22A shows the exhaust purification catalyst 13 when the hydrocarbon injection period is the optimal hydrocarbon injection period ΔT shown in FIG. 21B and the hydrocarbon injection time is the optimal hydrocarbon injection time WT shown in FIG. 21C. FIG. 22B shows a change in the air-fuel ratio detected by the downstream air-fuel ratio sensor 24 at this time. FIG. 22B shows the change in the air-fuel ratio when the downstream air-fuel ratio sensor 24 is not poisoned at all.
  22A and 22B, the rich-side peak air-fuel ratio (A / F) r of the air-fuel ratio (A / F) in of the exhaust gas flowing into the exhaust purification catalyst 13 is rich, whereas The rich-side peak air-fuel ratio (A / F) p of the air-fuel ratio detected by the downstream-side air-fuel ratio sensor 24 is lean. This is because part of the supplied hydrocarbons once adheres to the exhaust purification catalyst 13 and then evaporates with a time difference, thereby smoothing the change in the air-fuel ratio of the exhaust gas flowing out from the exhaust purification catalyst 13. This is considered to be because the peak value of becomes smaller.
  When the hydrocarbon injection period is the optimum hydrocarbon injection period ΔT shown in FIG. 21B and the hydrocarbon injection time is the optimum hydrocarbon injection time WT shown in FIG. The detected rich-side peak air-fuel ratio (A / F) p is obtained in advance by experiment, and this rich-side peak air-fuel ratio (A / F) p obtained in advance by experiment is determined as the target peak air-fuel ratio (A / F). / F) t is stored in advance for each engine operating state. In the embodiment of the present invention, this target peak air-fuel ratio (A / F) t is stored in advance in the ROM 32 in the form of a map as shown in FIG. 23 as a function of the engine speed N and the engine output torque Te. .
  Next, the structure of the upstream air-fuel ratio sensor 23 and the downstream air-fuel ratio sensor 24 used in the present invention will be briefly described. The upstream air-fuel ratio sensor 23 and the downstream air-fuel ratio sensor 24 have the same structure, and FIG. 24A schematically shows the structure of these air-fuel ratio sensors 23 and 24.
  Referring to FIG. 24A, the sensor unit 60 of the air-fuel ratio sensors 23, 24 includes a thin cup-shaped solid electrolyte 61 made of zirconia Zr, a platinum thin film electrode 62 that covers the inner peripheral surface of the solid electrolyte 61, and the solid electrolyte 61. A platinum thin film electrode 63 covering the outer peripheral surface and a diffusion resistance layer 64 made of alumina covering the periphery of the electrode 63 are configured. The sensor unit 60 is covered with a protective cover 66 having a large number of holes 65. The sensor unit 60 is disposed in the exhaust gas, and the exhaust gas is guided to the diffusion resistance layer 64 through the hole 65. As shown in FIG. 24A, a constant voltage E is applied between the electrodes 62 and 63. At this time, a current I as shown in FIG. 24B is applied between the electrodes 62 and 63 according to the air-fuel ratio of the exhaust gas. Flows. In the present invention, the air-fuel ratio is obtained from the current value I based on the relationship shown in FIG. 24B. That is, the air-fuel ratio is detected from the outputs of the air-fuel ratio sensors 23 and 24.
  FIG. 25A shows the first NO_XNO by purification method_XThe change in the air-fuel ratio AFR1 of the exhaust gas detected by the upstream air-fuel ratio sensor 23 and the change in the air-fuel ratio AFR2 of the exhaust gas detected by the downstream air-fuel ratio sensor 24 when the purification action is performed. Show. Since the upstream air-fuel ratio sensor 23 is located on the upstream side of the hydrocarbon supply valve 15, it is not affected by the hydrocarbon supply action from the hydrocarbon supply valve 15, and therefore, as shown in FIG. The air-fuel ratio AFR1 of the exhaust gas detected by the fuel ratio sensor 23 is the base air-fuel ratio (A / F) b.
  On the other hand, the air-fuel ratio AFR2 of the exhaust gas detected by the downstream air-fuel ratio sensor 24 changes accordingly when hydrocarbons are supplied from the hydrocarbon supply valve 15. That is, when hydrocarbon is supplied from the hydrocarbon supply valve 15, the air-fuel ratio AFR 2 detected by the downstream air-fuel ratio sensor 24 is a reference detected when no hydrocarbon is supplied from the hydrocarbon supply valve 15. The air-fuel ratio AFRB changes to the rich side as indicated by RS. Therefore, when hydrocarbons are supplied from the hydrocarbon supply valve 15 at an interval, the air-fuel ratio AFR2 detected by the downstream air-fuel ratio sensor 24 is an interval based on the reference air-fuel ratio AFRB as shown in FIG. 25A. Will change to the rich side.
  However, in this case, as shown in FIG. 25A, the air-fuel ratio is between the air-fuel ratio AFR1 detected by the upstream air-fuel ratio sensor 23 and the reference air-fuel ratio AFRB of the air-fuel ratio AFR2 detected by the downstream air-fuel ratio sensor 24. It has been found that the difference ΔAFR occurs, and the reference air-fuel ratio AFRB of the air-fuel ratio AFR2 detected by the downstream air-fuel ratio sensor 24 becomes richer than the air-fuel ratio AFR1 detected by the upstream air-fuel ratio sensor 23. The reason is considered as follows.
  That is, a hydrocarbon having a large molecular weight such as light oil is supplied from the hydrocarbon supply valve 15, and the hydrocarbon supplied from the hydrocarbon supply valve 15 is oxidized or partially oxidized in the exhaust purification catalyst 13. However, actually, not all hydrocarbons supplied from the hydrocarbon supply valve 15 are oxidized or partially oxidized, and some hydrocarbons pass through the exhaust purification catalyst 13 without being oxidized or partially oxidized. End up. Accordingly, the hydrocarbon having a high molecular weight that has passed through the exhaust purification catalyst 13 flows into the downstream air-fuel ratio sensor 24.
  By the way, in the air-fuel ratio sensor shown in FIG. 24A, oxygen and hydrocarbons contained in the exhaust gas diffuse in the diffusion resistance layer 64 to reach the electrode 63, and the hydrocarbons are oxidized on the electrode 63. At this time, if oxygen is excessive, oxygen ions move in the solid electrolyte 61 from the electrodes 63 to 62. If oxygen is insufficient, oxygen ions move in the solid electrolyte 61 from the electrodes 62 to 63. The air-fuel ratio is detected by the current I generated by the movement.
  On the other hand, as in the downstream air-fuel ratio sensor 24, when hydrocarbons having a large molecular weight flow into the air-fuel ratio sensor, some hydrocarbons diffuse into the diffusion resistance layer 64 and reach the electrode 63 as described above. Since some of the hydrocarbons have a large molecular weight, they are once deposited on the surface of the diffusion resistance layer 64. Next, the once deposited hydrocarbon having a high molecular weight gradually diffuses in the diffusion resistance layer 64 and reaches the electrode 63. As a result, the downstream air-fuel ratio sensor 24 determines that a large amount of hydrocarbons is still contained in the exhaust gas even after the hydrocarbons are supplied from the hydrocarbon supply valve 15. The reference air-fuel ratio AFRB of the detected air-fuel ratio AFR2 is richer than the air-fuel ratio AFR1 detected by the upstream air-fuel ratio sensor 23.
  In this case, the reference of the air-fuel ratio AFR1 detected by the upstream air-fuel ratio sensor 24 and the air-fuel ratio AFR2 detected by the downstream air-fuel ratio sensor 24 increases as the amount of hydrocarbon adhering to the downstream air-fuel ratio sensor 24 increases. The air-fuel ratio difference ΔAFR from the air-fuel ratio AFRB increases. On the other hand, the amount of hydrocarbon adhering to the downstream side air-fuel ratio sensor 24 increases as the amount of hydrocarbon passing through the exhaust purification catalyst 13 increases. Therefore, as shown in FIG. 25B, it can be seen that the air-fuel ratio difference ΔAFR increases as the amount of hydrocarbons WS that passes through the exhaust purification catalyst 13 increases. That is, the amount of hydrocarbons that pass through the exhaust purification catalyst 13 can be detected based on the air-fuel ratio difference ΔAFR.
  Therefore, in the present invention, the hydrocarbon supply valve 15 for supplying hydrocarbons is arranged in the engine exhaust passage, and the upstream side for detecting the air-fuel ratio of the exhaust gas in the engine exhaust passage upstream of the hydrocarbon supply valve 15. The air-fuel ratio sensor 23 is arranged, and NO contained in the exhaust gas in the engine exhaust passage downstream of the hydrocarbon supply valve 15_XAn exhaust purification catalyst 13 for reacting with the reformed hydrocarbon is disposed, and a downstream air-fuel ratio sensor 24 for detecting the air-fuel ratio of the exhaust gas is disposed in the engine exhaust passage downstream of the exhaust purification catalyst 13. The downstream air-fuel ratio sensor 24 has a solid electrolyte 61, electrodes 62 and 63 that cover both side surfaces of the solid electrolyte 61, and a diffusion resistance layer 64 that covers one electrode 63, and exhaust gas is a diffusion resistance layer 64. The noble metal catalyst 51, 52 is supported on the exhaust gas flow surface of the exhaust purification catalyst 13, and a basic exhaust gas flow surface portion 54 is provided around the noble metal catalyst 51, 52. The exhaust purification catalyst 13 vibrates the concentration of hydrocarbons flowing into the exhaust purification catalyst 13 with an amplitude within a predetermined range and a period within the predetermined range. NO contained in the exhaust gas_XNO in the exhaust gas if the vibration period of the hydrocarbon concentration is longer than a predetermined range._XTherefore, when the hydrocarbon is supplied from the hydrocarbon supply valve 15, the air-fuel ratio detected by the downstream air-fuel ratio sensor 24 is supplied from the hydrocarbon supply valve 15. The upstream side air-fuel ratio sensor 23 detects the amount WS of hydrocarbons that change to the rich side with respect to the reference air-fuel ratio AFRB detected when the engine is not, and pass through the exhaust purification catalyst 13 from the hydrocarbon feed valve 15. It is detected from the air-fuel ratio difference ΔAFR between the air-fuel ratio AFR1 thus made and the reference air-fuel ratio AFRB detected by the downstream air-fuel ratio sensor 24.
  In this case, as shown in FIG. 25A, the amount WS of hydrocarbons supplied from the hydrocarbon supply valve 15 and passing through the exhaust purification catalyst 13 increases as the air-fuel ratio difference ΔAFR increases.
  Now, the amount WS of hydrocarbons passing through the exhaust purification catalyst 13 increases due to various causes. For example, when the exhaust purification catalyst 13 deteriorates, the ability to partially oxidize or oxidize hydrocarbons decreases, so the amount of hydrocarbon WS that passes through the exhaust purification catalyst 13 increases. Further, even when the amount of hydrocarbons supplied from the hydrocarbon supply valve 15 is abnormally increased for some reason, the amount of hydrocarbons WS that passes through the exhaust purification catalyst 13 increases.
  Therefore, in the embodiment according to the present invention, when the amount WS of hydrocarbons supplied from the hydrocarbon supply valve 15 and passes through the exhaust purification catalyst 13 exceeds a predetermined limit value, whether the exhaust purification catalyst 13 has deteriorated, Alternatively, it is determined that the injection amount from the hydrocarbon supply valve 15 is abnormal.
  Next, two typical methods for detecting the air-fuel ratio difference ΔAFR will be described with reference to FIG. 26A.
  In the first method, a period I (FIG. 26A) in which the reference air-fuel ratio AFRB is detected is selected from the air-fuel ratios AFR2 detected by the downstream air-fuel ratio sensor 24. In this period I, the upstream air-fuel ratio sensor 23 is selected. This is a method for detecting the air-fuel ratio difference ΔAFR from the difference between the air-fuel ratio AFR1 detected by the above and the reference air-fuel ratio AFRB of the air-fuel ratio AFR2 detected by the downstream air-fuel ratio sensor 24 or the average value of these differences.
  In the second method, the air-fuel ratio detected by the upstream air-fuel ratio sensor 23 at a time St delayed by a time ft with respect to the timing of hydrocarbon injection from the hydrocarbon feed valve 15 indicated by a broken line Wt in FIG. 26A. This is a method of detecting the air-fuel ratio difference ΔAFR from the reference air-fuel ratio AFRB of the air-fuel ratio AFR2 detected by the AFR1 and the downstream air-fuel ratio sensor 24. In this case, the faster the exhaust gas flow rate, that is, the larger the intake air amount, the shorter the time ft during which the air-fuel ratio difference ΔAFR can be detected. This time ft is obtained in advance by experiments, and this time ft is stored in advance as a function of the intake air amount GA as shown in FIG. 26B.
  FIG. 27 shows an operation control routine for the overall operation of the engine. This routine is executed by interruption every predetermined time.
  Referring to FIG. 27, first, at step 60, the temperature TC of the exhaust purification catalyst 13 is determined from the output signal of the temperature sensor 25 to the activation temperature TC._OIt is discriminated whether or not it exceeds. TC <TC_OIn other words, when the exhaust purification catalyst 13 is not activated, the second NO_XIt is determined that the purification method should be used, and the process proceeds to step 61. In step 61, the discharge NO per unit time is determined from the map shown in FIG._XThe quantity NOXA is calculated. Next, at step 62, NO is discharged to ΣNOX._XOcclusion NO by adding the amount NOXA_XAn amount ΣNOX is calculated. Next, at step 63, NO is stored._XIt is determined whether or not the amount ΣNOX exceeds the allowable value MAX.
  When it is determined at step 63 that ΣNOX ≦ MAX, the routine proceeds to step 64 where fuel injection processing from the fuel injection valve 3 is performed. At this time, fuel is injected from the fuel injection valve 3 so as to obtain a predetermined lean air-fuel ratio determined from the operating state of the engine. On the other hand, when it is determined at step 63 that ΣNOX> MAX, the routine proceeds to step 65 where rich control I is performed. That is, the additional fuel amount WR is calculated from the map shown in FIG. 19, and the additional fuel injection action is performed. At this time, the NO stored from the exhaust purification catalyst 13_XIs released. Next, at step 66, ΣNOX is cleared.
  On the other hand, when it is judged at step 60 that TC ≧ TCo, that is, when the exhaust purification catalyst 13 is activated, the routine proceeds to step 67, where it is judged if TC <TCo at the previous interruption. When TC <TCo at the time of the previous interruption, that is, when the exhaust purification catalyst 13 is now activated, the routine proceeds to step 68 where the rich control II is executed. At this time as well, the additional fuel amount WR is calculated from the map shown in FIG._XIs released. Next, at step 69, ΣNOX is cleared.
  On the other hand, when TC ≧ TCo at the time of the previous interruption, that is, when the exhaust purification catalyst 13 has already been activated, the routine proceeds to step 70 where operation control I is executed. In this operation control I, the first NO according to the present invention._XNO by purification method_XPurifying action is performed. That is, when the exhaust purification catalyst 13 is not activated, the second NO_XNO by purification method_XWhen the purification action is performed and the exhaust purification catalyst 13 is activated, the second NO_XNO from the purification method_XSwitch to purification method.
  2nd NO_XNO from the purification method_XWhen switched to the purification method, NO is given to the exhaust purification catalyst 13_XIs stored, the NO stored in the exhaust purification catalyst 13_XIs released at once without being reduced. Therefore, in the example shown in FIG. 25, the NO stored in the exhaust purification catalyst 13 in this way._XIn order to prevent the NO from being released at once without being reduced, the second NO_XNO from the purification method_XImmediately before switching to the purification method, the second NO in step 68._XOccluded NO from the exhaust purification catalyst 13 by the purification method_XRich control II is performed to release the.
  Next, the operation control I performed in step 70, that is, the first NO_XNO by purification method_XThe purification action will be described. 28 and 29 show a first embodiment of the operation control I. In the first embodiment, the case where the air-fuel ratio difference ΔAFR is obtained by the first method described with reference to FIG. 26A is shown.
  Referring to FIG. 28, first, at step 80, the target value of the base air-fuel ratio, that is, the target air-fuel ratio (A / F) b is calculated from the map shown in FIG. 20B. Next, at step 81, the intake air amount GA is calculated from the output signal of the intake air amount detector 8. Next, at step 82, the fuel injection for generating output from the fuel injection valve 3 necessary for changing the base air-fuel ratio from the target air-fuel ratio (A / F) b and the intake air amount GA to the target air-fuel ratio (A / F) b. A quantity Q is calculated. Next, at step 83, the fuel injection time QT is calculated from the fuel injection amount Q, and then at step 84, fuel injection processing for injecting fuel from the fuel injection valve 3 according to the fuel injection time QT is performed.
  Next, at step 85, the optimum hydrocarbon injection cycle ΔT is calculated from the map shown in FIG. 21B. Next, at step 86, the optimum hydrocarbon injection time WT is calculated from the map shown in FIG. 21C. Next, at step 87, a hydrocarbon injection process for injecting hydrocarbons from the hydrocarbon supply valve 15 according to the hydrocarbon injection time WT is performed.
  Next, at step 88, the air-fuel ratio AFR1 detected by the upstream air-fuel ratio sensor 23 and the air-fuel ratio AFR2 detected by the downstream air-fuel ratio sensor 24 are read. Next, at step 89, the air-fuel ratio difference ΔAFR (= AFR1-AFR2) is calculated by subtracting the air-fuel ratio AFR2 detected by the downstream air-fuel ratio sensor 24 from the air-fuel ratio AFR1 detected by the upstream air-fuel ratio sensor 23. The Next, at step 90, in order to determine whether or not the air-fuel ratio AFR2 detected by the downstream air-fuel ratio sensor 24 is the reference air-fuel ratio AFRB, the air-fuel ratio difference ΔAFR between the previous interruption and the current interruption is determined. A change amount dAFR is calculated.
  Next, at step 91, it is judged if the absolute value of the change amount dAFR of the air-fuel ratio difference ΔAFR is smaller than a predetermined small value dX. Processing is performed when the absolute value of the change amount dAFR of the air-fuel ratio difference ΔAFR is larger than dX, that is, when the air-fuel ratio AFR2 detected by the downstream air-fuel ratio sensor 24 is changing rapidly as indicated by RS in FIG. 25A. Complete the cycle. On the other hand, when the absolute value of the change amount dAFR of the air-fuel ratio difference ΔAFR is smaller than dX, that is, when the fluctuation amount of the air-fuel ratio AFR2 detected by the downstream air-fuel ratio sensor 24 is small, the downstream air-fuel ratio sensor 24 It is determined that the detected air-fuel ratio AFR2 is the reference air-fuel ratio AFRB, and the routine proceeds to step 92.
  In step 92, the integrated value ΣAF of the air-fuel ratio difference ΔAFR is calculated by adding the air-fuel ratio difference ΔAFR to ΣAF. Next, at step 93, the count value N is incremented by one. Next, at step 94, it is judged if the count value N has reached a constant value No. When the count value N reaches the constant value No, the routine proceeds to step 95 where the integrated value ΣAF is divided by the constant value No to calculate the average value ΔAFRM of the air-fuel ratio difference ΔAFR. In this embodiment, the average value ΔAFRM of the air-fuel ratio difference ΔAFR is calculated in this way. Therefore, in this embodiment, the hydrocarbon amount WS that passes through the exhaust purification catalyst 13 can be calculated from the average value ΔAFRM of the air-fuel ratio difference ΔAFR. .
  When the average value ΔAFRM of the air-fuel ratio difference ΔAFR is calculated, the routine proceeds to step 96 where ΣAF and N are cleared. Next, at step 97, it is judged if the average value ΔAFRM of the air-fuel ratio difference ΔAFR is larger than a predetermined limit value Wo. When the average value ΔAFRM of the air-fuel ratio difference ΔAFR is larger than the limit value Wo, the routine proceeds to step 98, where an alarm is issued indicating that the exhaust purification catalyst 13 has deteriorated or that the hydrocarbon supply valve 15 is abnormal.
  Next, a second embodiment of the operation control 1 performed in step 70 of FIG. 27 will be described.
  Now, the first NO according to the present invention_XNO by purification method_XWhen the purification action is being performed, the air-fuel ratio of the exhaust gas discharged from the engine is the optimum base air-fuel ratio (A / F) b shown in FIG. 20B and is detected by the downstream air-fuel ratio sensor 24. When the rich peak air-fuel ratio (A / F) p of the air-fuel ratio is the target peak air-fuel ratio (A / F) t shown in FIG. 23, the maximum NOx purification rate is obtained. Therefore, in this second embodiment, based on the output signal of the upstream air-fuel ratio sensor 23, the fuel injection valve is set so that the air-fuel ratio of the exhaust gas discharged from the engine becomes the optimum base air-fuel ratio (A / F) b. 3 is feedback-controlled, and the rich peak air-fuel ratio (A / F) p of the air-fuel ratio detected by the downstream air-fuel ratio sensor 24 based on the output signal of the downstream air-fuel ratio sensor 24 is the target. The feed amount of hydrocarbons from the hydrocarbon feed valve 15 is feedback controlled so that the peak air-fuel ratio (A / F) t is obtained.
  More specifically, the final fuel injection from the fuel injection valve 3 is performed by multiplying the fuel injection amount QT calculated from the optimal base air-fuel ratio (A / F) b and the intake air amount GA by the correction coefficient G. The quantity QTO (= G · QT) is calculated, and the value of the correction coefficient G is feedback-controlled so that the air-fuel ratio of the exhaust gas discharged from the engine becomes the optimum base air-fuel ratio (A / F) b. That is, the value of the correction coefficient G necessary for setting the air-fuel ratio of the exhaust gas discharged from the engine to the optimum base air-fuel ratio (A / F) b is learned.
  Further, the final hydrocarbon injection time WTO (= K · WT) from the hydrocarbon feed valve 15 is calculated by multiplying the optimum hydrocarbon injection time WT calculated from the map of FIG. 21C by the correction coefficient K. Then, the value of the correction coefficient K is feedback controlled so that the rich peak air-fuel ratio (A / F) p of the air-fuel ratio detected by the downstream air-fuel ratio sensor 24 becomes the target peak air-fuel ratio (A / F) t. The That is, the value of the correction coefficient K necessary for setting the rich peak air-fuel ratio (A / F) p of the air-fuel ratio detected by the downstream air-fuel ratio sensor 24 to the target peak air-fuel ratio (A / F) t is learned. Like to do.
  By the way, the first NO_XNO by purification method_XIf the reference air-fuel ratio AFRB of the air-fuel ratio AFR2 detected by the downstream air-fuel ratio sensor 24 shifts to the rich side with respect to the air-fuel ratio AFR1 detected by the upstream air-fuel ratio sensor 23 when the purification action is being performed. Further, the rich-side peak air-fuel ratio (A / F) p of the air-fuel ratio detected by the downstream-side air-fuel ratio sensor 24 also shifts to the rich side. Therefore, if feedback control is performed so that the rich-side peak air-fuel ratio (A / F) p of the air-fuel ratio detected by the downstream air-fuel ratio sensor 24 at this time becomes the target peak air-fuel ratio (A / F) t, the hydrocarbons The amount of hydrocarbon injection from the supply valve 15 deviates from the optimum amount, and as a result, the NOx purification rate decreases.
  In this case, in order to prevent the NOx purification rate from decreasing, when the reference air-fuel ratio AFRB of the air-fuel ratio AFR2 detected by the downstream air-fuel ratio sensor 24 is shifted to the rich side, that is, the upstream air-fuel ratio sensor 23. When the air-fuel ratio difference ΔAFR between the air-fuel ratio AFR1 detected by the above and the reference air-fuel ratio AFRB of the air-fuel ratio AFR2 detected by the downstream air-fuel ratio sensor 24 exceeds a predetermined allowable value AX, for example, 0.3, It is necessary to prohibit the feedback control from being performed by the downstream air-fuel ratio sensor 24.
  Therefore, in the embodiment according to the present invention, the injection period of hydrocarbons from the hydrocarbon feed valve 15 is such that the concentration of hydrocarbons flowing into the exhaust purification catalyst 13 oscillates with a period within a predetermined range during engine operation. In addition to being controlled, the injection amount of hydrocarbons from the hydrocarbon feed valve 15 is controlled so that the amplitude of the change in the concentration of hydrocarbons flowing into the exhaust purification catalyst 13 becomes an amplitude within a predetermined range. When the difference ΔAFR is smaller than a predetermined allowable value AX, the output of the downstream side air-fuel ratio sensor 24 so that the amplitude of the change in the concentration of hydrocarbons flowing into the exhaust purification catalyst 13 becomes an amplitude within a predetermined range. Based on the signal, the injection amount of hydrocarbons from the hydrocarbon supply valve 15 is controlled, and when the air-fuel ratio difference Δ is larger than a predetermined allowable value AX, the output of the downstream air-fuel ratio sensor 24 is increased. Control of the injection amount of hydrocarbons from the hydrocarbon supply valve 15 based on the force signal is prohibited.
  In this case, as described above, when the air-fuel ratio difference ΔAFR is smaller than the predetermined allowable value AX, the downstream air-fuel ratio sensor 24 detects the amount of hydrocarbon injection from the hydrocarbon supply valve 15. Further, the value of the correction coefficient K necessary for setting the rich-side peak air-fuel ratio (A / F) p to the previously stored target air-fuel ratio (A / F) t, that is, a correction value is learned. In this case, even if the air-fuel ratio difference ΔAFR is larger than the predetermined allowable value AX, the control of the injection amount of hydrocarbons from the hydrocarbon supply valve 15 based on the output signal of the downstream air-fuel ratio sensor 24 is prohibited. When calculating the injection amount of hydrocarbons from the hydrocarbon supply valve 15, the value of this correction coefficient K, that is, the correction value is used, and accordingly, from the hydrocarbon supply valve 15 based on the output signal of the downstream air-fuel ratio sensor 24. When the control of the hydrocarbon injection amount is prohibited, the hydrocarbon injection amount from the hydrocarbon supply valve 15 is corrected using the correction coefficient K, that is, the learned correction value.
  On the other hand, if the period during which the hydrocarbon injection amount control is prohibited based on the output signal of the downstream air-fuel ratio sensor 24 becomes longer, it becomes necessary to relearn the value of the correction coefficient K. Next, a method for re-learning the value of the correction coefficient K will be described.
  FIG. 30A shows the first NO_XNO by purification method_XWhen the purification action is started and the first NO_XNO by purification method_XA change in the air-fuel ratio detected by the downstream air-fuel ratio sensor 24 when the purification action is stopped is shown. As shown at t1 in FIG. 30A, the first NO_XNO by purification method_XWhen the purification action is started, the reference air-fuel ratio AFRB of the air-fuel ratio AFR2 detected by the downstream air-fuel ratio sensor 24 is gradually richer away from the base air-fuel ratio (A / F) b, and at t2 in FIG. 30A First NO as shown_XNO by purification method_XWhen the purification action continues, the reference air-fuel ratio AFRB continues to shift to the rich side with respect to the base air-fuel ratio (A / F) b. On the other hand, the first NO_XNO by purification method_XWhen the purification action is stopped, the reference air-fuel ratio AFRB gradually increases toward the base air-fuel ratio (A / F) b as shown by t3 in FIG. 30A, and finally the base air-fuel ratio (A / F) ) B.
  In order to appropriately relearn the value of the correction coefficient K, the reference air-fuel ratio AFRB of the air-fuel ratio AFR1 detected by the upstream air-fuel ratio sensor 23 and the air-fuel ratio AFR2 detected by the downstream air-fuel ratio sensor 24 Needs to be made smaller than the allowable value AX. Therefore, in the embodiment according to the present invention, when the control of the injection amount of hydrocarbons from the hydrocarbon supply valve 15 based on the output signal of the downstream side air-fuel ratio sensor 24 is prohibited, the value of the correction coefficient K, that is, the correction value. Is determined to be re-learned, the hydrocarbon injection action from the hydrocarbon feed valve 15 is controlled so that the air-fuel ratio difference ΔAFR becomes smaller than a predetermined allowable value AX, thereby correcting the difference. The value of the coefficient K, that is, the correction value is relearned.
  As can be seen from FIG. 30A, the first NO_XNO by purification method_XWhen the purification action is stopped for t3 hours, the reference air-fuel ratio AFRB of the air-fuel ratio AFR2 detected by the downstream air-fuel ratio sensor 24 becomes the base air-fuel ratio (A / F) b, and therefore the air-fuel ratio detected by the upstream air-fuel ratio sensor 23 is increased. The air-fuel ratio difference ΔAFR between the air-fuel ratio AFR1 and the reference air-fuel ratio AFRB of the air-fuel ratio AFR2 detected by the downstream air-fuel ratio sensor 24 becomes smaller than the allowable value AX. Therefore, in one embodiment according to the present invention, when the value of the correction coefficient K is to be re-learned, the first NO_XNO by purification method_XThe purification action is stopped for t3 hours.
  In addition, as shown in FIG. 30B, when the value of the correction coefficient K is to be re-learned, the upstream air-fuel ratio sensor 23 detects it even if the hydrocarbon injection period from the hydrocarbon supply valve 15 is increased. The air-fuel ratio difference ΔAFR between the air-fuel ratio AFR1 and the reference air-fuel ratio AFRB of the air-fuel ratio AFR2 detected by the downstream air-fuel ratio sensor 24 becomes smaller than the allowable value AX. Therefore, in the embodiment according to the present invention, when the value of the correction coefficient K, that is, the correction value is to be re-learned, the hydrocarbon injection interval from the hydrocarbon feed valve 15 is temporarily increased as shown in FIG. 30B. Alternatively, as shown by t3 in FIG. 30A, the hydrocarbon injection from the hydrocarbon feed valve 15 is temporarily stopped.
  31 and 32 show a routine for executing the second embodiment of the operation control I. FIG. In the second embodiment, the air-fuel ratio difference ΔAFR is obtained by the second method described with reference to FIG. 26B, and the first NO is calculated when the value of the correction coefficient K is relearned._XNO by purification method_XThe case where the purification action is temporarily stopped is shown.
  Referring to FIG. 31, first, at step 100, the target value of the base air-fuel ratio, that is, the target air-fuel ratio (A / F) b is calculated from the map shown in FIG. 20B. Next, at step 101, the intake air amount GA is calculated from the output signal of the intake air amount detector 8. Next, at step 102, the fuel injection for generating output from the fuel injection valve 3 necessary to change the base air-fuel ratio from the target air-fuel ratio (A / F) b and the intake air amount GA to the target air-fuel ratio (A / F) b. A quantity Q is calculated. Next, at step 103, the fuel injection time QT is calculated from the fuel injection amount Q.
  Next, at step 104, the actual air-fuel ratio AFR1 of the exhaust gas discharged from the engine is detected based on the output of the upstream air-fuel ratio sensor 23. Next, at step 105, it is judged if the actual air-fuel ratio AFR1 is larger than a value obtained by adding a small constant value β to the target air-fuel ratio (A / F) b. When AFR1> (A / F) b + β, the routine proceeds to step 106, where the constant value ΔG is added to the correction coefficient G for the fuel injection time QT. Next, the routine proceeds to step 106, where the value (G · QT) obtained by multiplying the fuel injection time QT by the correction coefficient G is set as the final fuel injection time QTO.
  On the other hand, when it is determined in step 105 that AFR1> (A / F) b + β is not satisfied, the routine proceeds to step 107, where the actual air-fuel ratio AFR1 of the exhaust gas discharged from the engine is calculated from the target peak air-fuel ratio (A / F) b. It is determined whether or not the value is smaller than a value obtained by subtracting the constant value β. When AFR1 <(A / F) b−β, the routine proceeds to step 108, where the constant value ΔG is subtracted from the correction coefficient G, and the routine proceeds to step 109. Next, at step 110, a fuel injection process for injecting fuel from the fuel injection valve 3 according to the fuel injection time QTO is performed.
  Thus, in this embodiment, when AFR1> (A / F) b + β, the fuel injection time is increased, and when AFR1 (A / F) b−β, the fuel injection time is decreased, so that the exhaust gas discharged from the engine. Therefore, the actual base air-fuel ratio AFR1 is made to coincide with the target base air-fuel ratio (A / F) b.
  Next, at step 111, the optimum hydrocarbon injection cycle ΔT is calculated from the map shown in FIG. 21B. Next, at step 112, the optimum hydrocarbon injection time WT is calculated from the map shown in FIG. 21C. Next, at step 113, it is judged if the feedback control prohibition flag is set. This feedback control prohibition flag indicates that the air-fuel ratio difference ΔAFR between the air-fuel ratio AFR1 detected by the upstream air-fuel ratio sensor 23 and the air-fuel ratio AFR2 detected by the downstream air-fuel ratio sensor 24 is greater than the allowable value AX. When it is small, it is reset.
  When the feedback control prohibition flag is reset, that is, when the air-fuel ratio difference ΔAFR is smaller than the allowable value AX, the routine proceeds to step 114 where the target peak air-fuel ratio (A / F) t is calculated from the map shown in FIG. . Next, at step 115, the actual rich peak air-fuel ratio (A / F) p is detected from the output of the downstream air-fuel ratio sensor 24.
  Next, at step 116, it is judged if the actual rich side peak air-fuel ratio (A / F) p is larger than a value obtained by adding a small constant value α to the target peak air-fuel ratio (A / F) t. When (A / F) p> (A / F) t + α, the routine proceeds to step 117, where a constant value ΔK is added to the correction coefficient K for the hydrocarbon injection time WT. Next, the routine proceeds to step 120, where the value (K · WT) obtained by multiplying the hydrocarbon injection time WT by the correction coefficient K is made the final hydrocarbon injection time WTO.
  On the other hand, when it is determined in step 116 that (A / F) p> (A / F) t + α is not established, the routine proceeds to step 118, where the actual rich-side peak air-fuel ratio (A / F) p becomes the target peak air-fuel ratio (A / F) It is determined whether or not the value is smaller than a value obtained by subtracting a constant value α from t. When (A / F) p <(A / F) t−α, the routine proceeds to step 119, where the constant value ΔK is subtracted from the correction coefficient K, and the routine proceeds to step 120. Next, at step 121, a hydrocarbon injection process for injecting hydrocarbons from the hydrocarbon supply valve 15 according to the final hydrocarbon injection time WTO is performed.
  Thus, when (A / F) p> (A / F) t + α, the hydrocarbon injection time is increased, and when (A / F) p <(A / F) t−α, the hydrocarbon injection time is decreased. Therefore, the actual rich-side peak air-fuel ratio (A / F) p is matched with the target peak air-fuel ratio (A / F) t.
On the other hand, when it is determined in step 113 that the feedback control prohibition flag is set, the routine jumps to step 120, where the final hydrocarbon injection time WTO (= K · WT) is calculated. At this time, feedback control of the hydrocarbon injection amount from the hydrocarbon supply valve 15 based on the output signal of the downstream air-fuel ratio sensor 24 is prohibited, and the final hydrocarbon injection is performed using the learned correction coefficient K value. Time WTO is calculated.
  FIG. 33 shows a routine for controlling the feedback control prohibition flag. This routine is executed by interruption every predetermined time.
  Referring to FIG. 33, first, at step 130, it is judged if time ft has passed since the hydrocarbon injection from the hydrocarbon feed valve 15 was performed. This time ft is obtained from the relationship shown in FIG. 26B. When the time ft has not elapsed since the hydrocarbon injection from the hydrocarbon feed valve 15 was performed, the processing cycle is completed. On the other hand, when time ft has elapsed since the hydrocarbon injection operation from the hydrocarbon feed valve 15 was performed, the routine proceeds to step 131 where the air-fuel ratio AFR1 and the downstream air-fuel ratio detected by the upstream air-fuel ratio sensor 23 are detected. The air-fuel ratio AFR2 detected by the sensor 24 is read. Next, at step 132, the air-fuel ratio difference ΔAFR (= AFR1−AFR2) is calculated by subtracting the air-fuel ratio AFR2 detected by the downstream air-fuel ratio sensor 24 from the air-fuel ratio AFR1 detected by the upstream air-fuel ratio sensor 23. The At this time, the air-fuel ratio AFR2 detected by the downstream air-fuel ratio sensor 24 represents the reference air-fuel ratio AFRB.
  Next, at step 133, the air-fuel ratio difference ΔAFR between the air-fuel ratio AFR1 detected by the upstream air-fuel ratio sensor 23 and the air-fuel ratio AFR2 detected by the downstream air-fuel ratio sensor 24 is smaller than the allowable value AX. Is determined. When the air-fuel ratio difference ΔAFR is smaller than the allowable value AX, the routine proceeds to step 134 where the feedback control prohibition flag is reset, and then the processing cycle is completed. Therefore, at this time, the injection amount of hydrocarbons from the hydrocarbon supply valve 15 is feedback-controlled based on the air-fuel ratio AFR2 detected by the downstream air-fuel ratio sensor 24.
  On the other hand, the routine proceeds to step 135 where it is determined that the air-fuel ratio difference ΔAFR in step 133 is larger than the allowable value AX, and the feedback control prohibition flag is set. Therefore, at this time, the feedback control of the hydrocarbon injection amount from the hydrocarbon supply valve 15 by the downstream air-fuel ratio sensor 24 is stopped. Next, the routine proceeds to step 136, where it is judged if it is time to relearn the value of the correction coefficient K. For example, when a certain amount of time has elapsed since the feedback control was prohibited and the vehicle is in a driving state suitable for learning the value of the correction coefficient K, or the vehicle has traveled more than a certain distance and corrected after the feedback control is prohibited. It is determined that it is time to relearn the value of the correction coefficient K when the driving state is suitable for learning the value of the coefficient K.
  When it is determined in step 136 that it is not time to re-learn the value of the correction coefficient K, the processing cycle is completed. On the other hand, when it is determined at step 136 that it is time to re-learn the value of the correction coefficient K, the routine proceeds to step 137, where hydrocarbons from the hydrocarbon feed valve 15 are discharged for a period indicated by t3 in FIG. 30A. Supply is stopped. When the supply of hydrocarbons from the hydrocarbon supply valve 15 is stopped in this way, the reference air of the air-fuel ratio AFR 1 detected by the upstream air-fuel ratio sensor 23 and the air-fuel ratio AFR 2 detected by the downstream air-fuel ratio sensor 24. The air-fuel ratio difference ΔAFR with the fuel ratio AFRB becomes smaller than the allowable value AX, and the feedback control prohibition flag is reset. As a result, the value of the correction coefficient K is relearned. When the value of the correction coefficient K is to be re-learned, as described above, the hydrocarbon injection period from the hydrocarbon supply valve 15 can be temporarily extended as shown in FIG. 30B.
  As another embodiment, an oxidation catalyst for reforming hydrocarbons may be disposed in the engine exhaust passage upstream of the exhaust purification catalyst 13.

DESCRIPTION OF SYMBOLS 4 ... Intake manifold 5 ... Exhaust manifold 7 ... Exhaust turbocharger 12a, 12b ... Exhaust pipe 13 ... Exhaust purification catalyst 14 ... Particulate filter 15 ... Hydrocarbon supply valve 23 Upstream air-fuel ratio sensor 24 Downstream air-fuel ratio sensor

Claims (12)

A hydrocarbon supply valve for supplying hydrocarbons is arranged in the engine exhaust passage, an upstream air-fuel ratio sensor for detecting the air-fuel ratio of the exhaust gas is arranged in the engine exhaust passage upstream of the hydrocarbon supply valve, An exhaust purification catalyst for reacting NO _X contained in the exhaust gas with the reformed hydrocarbon is disposed in the engine exhaust passage downstream of the hydrocarbon supply valve, and the engine exhaust passage downstream of the exhaust purification catalyst is disposed in the engine exhaust passage. A downstream air-fuel ratio sensor for detecting the air-fuel ratio of the exhaust gas is disposed, the downstream air-fuel ratio sensor includes a solid electrolyte, electrodes that respectively cover both sides of the solid electrolyte, and a diffusion resistance layer that covers one electrode And a noble metal catalyst is supported on the exhaust gas flow surface of the exhaust purification catalyst and a basic exhaust around the noble metal catalyst. Gas distribution surface When the concentration of hydrocarbons flowing into the exhaust purification catalyst is vibrated with an amplitude within a predetermined range and a period within the predetermined range, the exhaust purification catalyst which has a property for reducing the NO _X contained, and the vibration period of the hydrocarbon concentration has a property of absorbing the amount of NO _X contained in the exhaust gas to be longer than the range defined the advance is increased The air-fuel ratio detected by the downstream air-fuel ratio sensor when hydrocarbon is supplied from the hydrocarbon supply valve is richer than the reference air-fuel ratio detected when no hydrocarbon is supplied from the hydrocarbon supply valve. The amount of hydrocarbons supplied from the hydrocarbon supply valve and passing through the exhaust purification catalyst is determined based on the air-fuel ratio detected by the upstream air-fuel ratio sensor and the reference air-fuel detected by the downstream air-fuel ratio sensor. Exhaust purification system of an internal combustion engine so as to detect the air-fuel ratio difference between.   The exhaust gas purification apparatus for an internal combustion engine according to claim 1, wherein the amount of hydrocarbons supplied from the hydrocarbon supply valve and passing through the exhaust gas purification catalyst increases as the air-fuel ratio difference increases.   During the engine operation, the hydrocarbon injection period from the hydrocarbon supply valve is controlled so that the concentration of hydrocarbons flowing into the exhaust purification catalyst oscillates with a period within the predetermined range. The amount of hydrocarbon injection from the hydrocarbon feed valve is controlled so that the amplitude of the concentration change of the inflowing hydrocarbon becomes an amplitude within the predetermined range, and the air-fuel ratio difference is more than a predetermined allowable value. Is smaller, the hydrocarbon concentration from the hydrocarbon supply valve is controlled based on the output signal of the downstream air-fuel ratio sensor so that the amplitude of the change in the concentration of the hydrocarbon flowing into the exhaust purification catalyst becomes an amplitude within a predetermined range. When the injection amount is controlled and the air-fuel ratio difference is larger than a predetermined allowable value, control of the hydrocarbon injection amount from the hydrocarbon supply valve based on the output signal of the downstream air-fuel ratio sensor is prohibited. An exhaust purification system of an internal combustion engine according to claim 2 that.   When the air-fuel ratio difference is smaller than a predetermined allowable value, the rich-side peak air-fuel ratio detected by the downstream air-fuel ratio sensor is stored in advance for the amount of hydrocarbon injection from the hydrocarbon feed valve. A correction value necessary to obtain the target air-fuel ratio is learned, and the difference in air-fuel ratio becomes larger than a predetermined allowable value, so that the carbonization from the hydrocarbon supply valve based on the output signal of the downstream air-fuel ratio sensor is performed. The exhaust emission control device for an internal combustion engine according to claim 3, wherein when the control of the hydrogen injection amount is prohibited, the hydrocarbon injection amount from the hydrocarbon supply valve is corrected using the learned correction value.   When it is determined that the correction value should be re-learned when control of the injection amount of hydrocarbons from the hydrocarbon supply valve based on the output signal of the downstream air-fuel ratio sensor is prohibited, the air-fuel ratio is determined. The internal combustion engine according to claim 4, wherein the hydrocarbon injection from the hydrocarbon feed valve is controlled so that the difference becomes smaller than a predetermined allowable value, thereby re-learning the correction value. Exhaust purification device.   When the correction value is to be re-learned, the hydrocarbon injection interval from the hydrocarbon supply valve is temporarily extended, or the hydrocarbon injection action from the hydrocarbon supply valve is temporarily stopped. 6. An exhaust emission control device for an internal combustion engine according to 5.   The amount of fuel supplied to the engine combustion chamber is controlled based on the output signal of the upstream air-fuel ratio sensor so that the air-fuel ratio of the exhaust gas discharged from the engine becomes a predetermined air-fuel ratio. 3. An exhaust emission control device for an internal combustion engine according to 3. In the exhaust purification catalyst, NO _X contained in the exhaust gas reacts with the reformed hydrocarbon to produce a reducing intermediate containing nitrogen and hydrocarbons, and the injection cycle of the hydrocarbon has a reducing intermediate. The exhaust emission control device for an internal combustion engine according to claim 3, wherein the exhaust gas purification period is a period necessary for continuously generating the body.   The exhaust gas purification apparatus for an internal combustion engine according to claim 8, wherein the hydrocarbon injection period is between 0.3 seconds and 5 seconds.   When the amount of hydrocarbons supplied from the hydrocarbon supply valve and passing through the exhaust purification catalyst exceeds a predetermined limit value, the exhaust purification catalyst has deteriorated or the injection amount from the hydrocarbon supply valve is abnormal The exhaust gas purification apparatus for an internal combustion engine according to claim 2, wherein the exhaust gas purification apparatus is determined.   The exhaust purification device for an internal combustion engine according to claim 1, wherein the noble metal catalyst is composed of platinum Pt and at least one of rhodium Rh and palladium Pd. The basic layer comprising a metal which can donate electrons to the alkali metal or alkaline earth metal or rare earth or NO _X in the exhaust gas flow on the surface of the exhaust purification catalyst is formed, the surface of the base layer is the base The exhaust gas purification apparatus for an internal combustion engine according to claim 1, wherein the exhaust gas circulation surface portion is formed.

Images